ChatChat Models
MiniMax
MiniMax M2-her
Best Overall
Built for dialogue, role-play, and multi-turn conversations.
DialogueRole-play
Same provider
More from MiniMax
MiniMax
MiniMax M3
Frontier multimodal coding model with a 1M context window.
MiniMax
MiniMax M2.7
Beginning the journey of recursive self-improvement at roughly 60 tokens per second.
MiniMax
MiniMax M2.7 Highspeed
Same M2.7 performance, faster and more agile at roughly 100 tokens per second.
MiniMax
MiniMax M2.5
Peak performance and ultimate value for mastering complex work at roughly 60 tokens per second.